Apart from non-melanoma skin cancer, prostate cancer is the most commonly diagnosed cancer among men in the United States today. In spite of its importance, few modifiable predictors of this disease have been established. Well-established risk factors for prostate cancer, such as age, family history, race, and country of residence are not amenable to modification and, hence, have limited utility as targets of primary prevention strategies. Several modifiable characteristics have been proposed to predict risk of prostate cancer. These include physical activity, various anthropometric measures, cigarette smoking, alcohol consumption, and diet. However, data regarding these associations have been inconsistent. To provide more information, we propose to analyze the associations of these characteristics with prostate cancer risk using data that already are available. In 1988, 11,122 men, with mean age 68 years and who were free of cancer, returned a health questionnaire. Men provided a wide range of information, including data on physical activity, weight, height, waist girth, smoking, alcohol intake, diet, and vitamin supplements. Repeated assessments of these characteristics also have been made over a span of up to 26 years prior to 1988, on questionnaires in 1962 or 1966 and 1977. In addition, men had undergone a standardized medical examination when they were aged 18 years. During follow-up from 1988 through 1993, 495 cases of prostate cancer developed. This large database provides a unique opportunity, at relatively little cost, to clarify the associations of various modifiable characteristics with prostate cancer risk. Because repeated assessments are available, changes and continuities can be examined in relation to risk. Findings from these analyses, especially with regard to changes, will have public health significance.